A Collective Anomaly Detection Technique to Detect Crypto Wallet Frauds on Bitcoin Network

Abstract

The popularity and remarkable attractiveness of cryptocurrencies, especially Bitcoin, absorb countless enthusiasts every day. Although Blockchain technology prevents fraudulent behavior, it cannot detect fraud on its own. There are always unimaginable ways to commit fraud, and the need to use anomaly detection methods to identify abnormal and fraudulent behaviors has become a necessity. The main purpose of this study is to use the Blockchain technology of symmetry and asymmetry in computer and engineering science to present a new method for detecting anomalies in Bitcoin with more appropriate efficiency. In this study, a collective anomaly approach was used. Instead of detecting the anomaly of individual addresses and wallets, the anomaly of users was examined. In addition to using the collective anomaly detection method, the trimmed_Kmeans algorithm was used for clustering. The results of this study show the anomalies are more visible among users who had multiple wallets. The proposed method revealed 14 users who had committed fraud, including 26 addresses in 9 cases, whereas previous works detected a maximum of 7 addresses in 5 cases of fraud. The suggested approach, in addition to reducing the processing overhead for extracting features, detect more abnormal users and anomaly behavior.

Anomaly Detection in Blockchain Networks Using Unsupervised Learning: A Survey

“…Cluster analysis has also been used in supporting consensus protocols [152][153][154][155][156][157][158][159][160] For example, Khenfouci et al [125], to avoid data tampering and fraudulent activities, developed a customized clustering-based consensus protocol to carry out a decentralized consensus mechanism, according to which the k-Means was applied locally by multiple competitive miners. The methodology comprised four layers (i.e., data layer, network layer, blockchain layer, and machine learning layer) and had two main actors: management and miner.…”

“…The final number of clusters was determined through optimal clustering, where the sum of square distances within clusters played the role of the performance index. The trimmed k-Means was also used in [160] to develop a collective anomaly detection approach, diverging from the conventional methods in that instead of implementing anomaly detection considering individual addresses and wallets, the study focused on scrutinizing anomalies at the user level, where the available dataset was taken from Kondor et al [161]. An interesting result of the study indicated that anomalies were more conspicuous among users with multiple wallets.…”

Illicit Activity Detection in Bitcoin Transactions using Timeseries Analysis

“…Many techniques have been developed for bitcoin fraud detection apart from the usage of timestamp data such as clustering [8], various techniques such as trimmed k-means, DBSCAN etc. are tested to this end.…”

Detecting anomalous cryptocurrency transactions: An AML/CFT application of machine learning-based forensics

“…Even if the ledger transparency featured by public blockchains mitigates the risk of fraudulent behavior, the technology is vulnerable to unpredictable exploitation methods (Shayegan et al, 2022;Xu, 2016). This prompted the development of specific techniques of anomaly detection.…”

“…The goal is to single out rare or suspicious events/items-i.e., those significantly different from the dataset (Kamišalić et al, 2021). While collective anomaly detection methods target groups of data points that differ from most of the data, point anomaly detection also considers single data points (Li et al, 2022;Shayegan et al, 2022). AML/CFT-regulated entities, especially in the financial industry, deploy RegTech solutions to screen their operations and detect anomalous activities in an automated way.…”